Skip to main content

Full text of "Ringed impact craters on Venus: An analysis from Magellan images"

See other formats

2 International Colloquium on Venus 


-i 1 r- 



1« 12 10 8 6 


Fig. 1. Contours of brightness ratio (data/model) of the 13a n m emission 
displaying the local-time asymmetry (see text). The brightness ratio patterns 
can be thought of as anomalous patterns in O. (a) and (b) are derived from two 
different types of PVOUVS data: (■) contains more data, so it better displays 
the northern hemisphere pattern, but in (b) we can see to more southerly 
latitudes. The northern hemisphere pattern is mirrored there. 

westward afternoon winds than with the eastward morning winds. 
Figure 2 shows some example eddy diffusion profiles calculated 
with this method. Early morning a.m. values are typically a factor 
of 10 or more larger than corresponding p.m. values. These local- 
time asymmetries will be discussed in more detail in the context of 
several scenarios for the middle atmosphere wave forcing. This 
same mechanism can also give rise to asymmetric wave drag forces 
that have the potential for generating the thermospheric superrotation. 
The mechanism for in situ forcing of the thermospheric superrotation 
is still considered a mystery, so this corollary provides strong 
support to our cloud-level wave source hypothesis. 

Fig. 2. Eddy diffusion coefficients characteristic of 8 a.m. and 4 p.m. wind 
profiles derived with the Lindzen parameterization [7]. Also shown are the 
VTGCM and von Zahn et al. [8] one-dimensional photochemical model AT for 

References: [1] Bougher et al. (1988) Icarus, 73, 545-573. 
[2] Keating etal. (1986) Adv. Space Res.. 5. 117-171. [3] Hedinet 
al. (1983) JGR, 88, 73-S3. [4] Niemann et al. (1980) JGR 85 
7817-7827. [5] Meier R. R. and Lee J.-S. (1982) Planet. Space Sci'., 
30, 439-450. [6] Alexander et al. (1991) Bull. AAS, 23, 1194 
[7] Lindzen R. S. (1981) JGR, 86, 9707-9714. f8] von Zahn et al 

(1980) JGR. 85, 7829-7840. 


M U H 


FROM MAGELLAN IMAG ES. Jim S. Alexopoulos and William 
B. McKinnon, Department of Earth and Planetary Sciences and 
McDonnell Center for the Space Sciences, Washington University 
St. Louis MO 63130, USA. 

We have analyzed cycle 1 Magellan images covering -90% of 
the venusian surface and have identified 55 unequivocal peak-ring 
craters and multiringed impact basins. This comprehensive study 
(52 peak-ring craters and at least 3 multiringed impact basins) 
complements our earlier independent analysis of Arecibo and 
Venera images and initial Magellan data [1,2] and that of the 
Magellan team [3]. 

Peak-ring craters are characterized by an outer, well-defined 
radar-bright rim, and an inner bright ring defined by a concentric 
arrangement of isolated and sometimes clustered peaks. The general 
appearance of venusian peak-ring craters, including their innerrings 
and crater rims, is morphologically similar to equivalent structures 
on the Moon, Mars, and Mercury. Some venusian peak rings, 
however, are distorted in shape (noncircular outline of inner ring) 
and off-center. Ejecta morphologies around peak-ring craters vary 
from regular and symmetric, including those due to oblique impact, 
to irregular and asymmetric. Many peak -ring craters exhibit out- 
flow channels. The outflows emanate from within the crater and 
breach the crater rim (e.g., Cleopatra), while others extend away 
from the distal end of the ejecta blanket or have been incorporated 
within the main ejecta deposit (e.g., Cochran). The interiors of most 
peak-nng craters are radar-dark or smooth (similar to surrounding 
plains), although some are bright A few larger craters are com- 
pletely flooded and show no interior structure or inner ring (e.g., 
Koidula, -70 km in diameter, and Alcolt, -65 km in diameter). The 
radar-smooth signature of the interior is likely due to postimpact 
resurfacing, either volcanism or (possibly differentiated) impact 

LP! Contribution No. 789 3 

melt. Many peak-ring craters also exhibit radar-bright returns 
associated with rough material around peaks. Numerous central- 
peak structures and peak-ring craters also show darker regions at the 
periphery of the crater floor. These are probably areas of localized 
flooding, or possibly pools of impact melt. A few structures also 
exhibit radar-bright regions near the periphery of the floor that may 
represent slumped material from thecraterrim. Fractures (orridges) 
associated with some structures (i.e., Isabella, -1 70 km in diameter, 
and Mona Lisa, -8 1 km in diameter) exhibit a radial and concentric 
pattern interior to the inner ring. This tectonic fabric is the possible 
manifestation of stresses associated with viscous relaxation. 

A plot of crater-rim to peak-ring diameter ratio against crater 
diameter (Fig. 1) shows that this ratio is relatively large (-5) at the 
transition diameter from central-peak craters and declines to less 
than 2 at larger diameters. The onset diameter to peak-ring forms on 
Venus is -40 km [1 ], although peak -ring craters as small as ~30 km 
in diameter have now been identified (Fig. 1). The inner rings of 
these smaller structures are usually comprised of small, isolated, 
and concentrically arranged peaks, and thus are not as distinct as the 
coherent ring mountains of larger peak-ring craters. These smaller 
peak-ring forms represent transitional forms between complex 
craters with central and multiple peaks and well-developed peak- 
ring craters. The fact that ratios are larger at the transition and that 
ring ratios decrease with crater size is consistent with peak -ring 
crater formation being an extension of the process of central peak 
formation. Specifically, peak rings may form by hydrodynamic 
uplift and subsequent collapse of an increasingly unstable central 
peak. Upon collapse, excess material may be redistributed around 
the central collapse area, forming a cluster of multiple peaks or a 
small ring of concentrically arranged peaks. At larger scales, 
complete peak collapse may allow for a relatively wider redistribu- 
tion of material, leading to smaller ring diameter ratios. Lending 
support to this argument may be the radar-bright returns associated 
with, and sometimes surrounding, theconcentrically arranged peaks 
and with the central peak or peak complexes of larger complex 
craters. This brighter zone may be rougher material (at radar 
wavelength scales) associated with the central peak collapse pro- 
cess. On the other hand, no craters with both a peak ring and central 
peak have been identified, in contrast with the Moon, Mercury, and 
Mars. The scatter in ring ratios at a given crater diameter (Fig. 1 ) is 
real. A possible explanation is that the degree of hydrodynamic 
collapse of the central peak and subsequent modification, for a given 
crater size, depends on differences in regional postshock target rock 
properties (specifically, the effective viscosity and yield strength). 

Q 3 


i ■ ■ ■ • i • 

• • 



■ ;*. . 

• • 



100 ISO 200 S50 


The declining ring ratio with diameter trend in Fig. 1 is similar 
to that observed for peak-ring craters on Mercury [ 1 ,2], albeit over 
a smaller-scale range. Most of the larger ringed craters with ratios 
of -2 or less are similar in appearance to the smaller peak -ring 
forms. They exhibit an inner ring of isolated and concentrically 
arranged massifs (e.g., Marie Celeste, -97 km in diameter, and 
Bonnevie, -84 km in diameter), i.e., they have a distinct peak ring. 
However, others (e.g., Mona Lisa) exhibit very bright radar returns 
from continuous ridgelike portions of the inner ring that may 
actually be scarps. Scarplike inner rings may represent a new class 
of impact structure, intermediate between peak-ring craters and true 
multiringed basins, or the degree of floor uplift may simply be 
pronounced for the largest peak-ring craters . The fact that the ejecta 
deposits extend away from the rim of Mona Lisa means that the 
inner ring probably formed within the present crater run. This is 
unlike venusian multiringed basins, where ejecta deposits can be 
seen to extend away from the first ring interior to the outer ring, 
implying that the outer ring formed outside what would approxi- 
mate the crater rim in a smaller complex crater. 

The transition from central-peak to peak-ring forms occurs at a 
range of diameters on the terrestrial planets that are determined 
mainly by surface gravity and postshock material properties [2,4], 
The observed average onset diameter of -40 km is consistent with 
our previous study [2] and, along with the transitions on the other 
terrestrial planets, supports our conclusion that the effective viscos- 
ity of cratered rubble during the modification stage is linearly 
proportional to crater diameter. 

Of the four largest ringed craters yet identified on Venus, three 
appear to be multiringed impact basins. Klenova (-144 km in 
diameter), LiseMeimer(~148km in diameter), and Mead (-270 km 
in diameter) are morphologically different from peak -ring craters, 
but similar to larger multiringed basin forms on the Moon, specifi- 
cally Orientale. Our initial classification [1,2] of Klenova as a 
multiringed basin was based on the scarplike appearance of its outer 
rings as identified in Venera images. The Magellan image of 
Klenova confirms our initial interpretation. Klenova exhibits three 
distinct rings: an inner ring of concentrically arranged peaks (peak- 
ring) analogous to Orien tale's Inner Rook, and two outerrings (main 
and intermediate) that we interpret to be possible fault scarps similar 
to Orientale's Cordillera ring and Outer Rook respectively. The 
ejecta blanket, not as evident in Venera images, is distinct and 
symmetrical and shows fields of secondary craters beyond the 
continuous ejecta deposits. 

The Arecibo image of Lise Meitner exhibits two distinct radar- 
bright rings that we interpret to be scarplike and thus similar to 
Orientale's Cordillera ring and Outer Rook. The interior of Meitner 
is uniformly radar dark, implying a likely smooth floor of postimpact 
volcanic deposits. Meitner also exhibits an outer, radar-bright 
feature, which is at the correct spacing to be a possible partial ring, 
and a feature, -9 km in width, interior to the inner ring that may be 
another ring or, more likely, a terrace similar to that produced by 
slumping of complex craters (Fig. 2). Analysis of the cycle 2 
Magellan image of Meitner should better reveal the nature of these 
structural elements. 

The largest impact structure identified on Venus, Mead, is 
comprised of two distinct rings. The radar-bright returns associated 
with each ring and an initial topographic evaluation indicate that the 
two rings are possible fault scarps analogous to Orientale's Cordil- 
lera ring and Outer Rook. The radar appearance of the ejecta 
deposits around Mead is suppressed relative to and not as distinct as 
that of peak -ring craters, but ejecta do occur between the two rings 
and extend across the outer one. The uniform radar returns associ- 

4 International Colloquium on Venus 

75 km 

|?3 Outer or Transitional Ring malarial 

■ Main Ring malar III 

[]J{] Innar Ring malarial 

[vjvj B«nch malarial 

El E|.cl» 

|"~| Interior and surrounding plains 

Fig. 2. 

ated with Mead's interior probably indicate flooding of the topo- 
graphically lower and relatively flat basin floor, with fracturing 
(mainly toward the center) producing locally radar-bright returns. 

Isabella, with a diameter of -170 km, is a ringed crater of the 
same scale, but its interior has been extensively flooded. A partially 
concentric arrangement of isolated peaks defines theremnants of an 
inner ring and a possible intermediate ring, both within the well- 
defined crater rim. The diameter ratio of the crater rim to the inner 
ring of peaks is roughly 2. If the intermediate ring, or a remnant 
thereof, can be shown to be scarplike in nature, then Isabella would 
also be a multiringed basin as interpreted here. 

The ring diameter ratios of the three unequivocal multiringed 
impact basins are distinctly different from peak-ring craters (Fig. 1 ), 
although they follow the trend of decreasing ring ratios with 
increasing diameter. The ring diameter ratios for the two most 
distinct rings for Klenova, Meitner, and Mead are -1.6, -1.6, and 
-1 .4 respectively. Also, ring ratios for Klenova's intermediate ring 
to peak ring is -1 .4, as is Meitner's partial ring to main ring. These 
ring ratios are close 10 the -Jl ratio suggested for Orientale and 
other lunar mulliring basins, thus supporting the multiringed basin 
analogy. Finally, theoretical arguments [1 2] support the formation 
of multiringed basins on Venus at these scales (>10O-150-km 

References: [1] Alexopoulos J. S. et al. (1991) LPSC XXII, 
13-14. [2] Alexopoulos J. S. and McKinnon W. B. (1992) Icarus, 
submitted. [3]SchaberG.G.etal.(1992)/G/?, submitted. [4] Melosh 
H J (1989) Impact Cralering: A Geologic Process, Oxford. New 


Jafar Arkani-Hamed, Department of Geological Sciences, McGill 
University, Montreal, Canada, H3A 2A7. 

The mountain belts of Ishtar Terra and the surrounding tesserae 
are interpreted as compressional regions [1,2,3]. The gravity and 
surface topography of western Ishtar Terra suggest a thick crust of 
60-110 km [4,5] that results from crustal thickening through 
tectonic processes. Underthrusting was proposed for the regions 
along Danu Montes [6] and Itzpapalotl Tessera [7]. Crustal thicken- 
ing was suggested for the entire Ishtar Terra [8]. In this study, three 
lithospheric models with total thicknesses of 40, 75, and 120 km and 
initial crustal thicknesses of 3, 9, and 18 km are examined. These 
models could be produced by partial melting and chemical differen- 
tiation in the upper mantle of a colder, an Earth-like, and a hotter 
Venus having temperatures of respectively 130O°C, 1400°C, and 
1500°C at the base of their thermal boundary layers associated with 
mantle convection. The effects of basalt-granulite-eclogite trans- 

formation (BGET) on the surface topography of a thickening 
basaltic crust is investigated adopting the experimental phase 
diagram [9] and density variations through the phase transformation 

Figure la shows the thermal evolution of the lithosphere of the 
cold Venus model with a linear crustal thickening of 0.5 km/m.y. 
followed by an exponential thickening for only 20 m.y. starting at 
100 m.y. with a characteristic time of 20 m.y. (the main results are 
not very sensitive to these values; see below). Figure lb shows the 
stability field of different phases that basalt enters. The BGET 
begins when the crust reaches a thickness of 7 km, and eclogite 
appears when the crust thickens beyond 70 km. Geologically 
speaking, the BGET is assumed to be instantaneous. Ahren and 
Schubert [11] suggested that cold basalt may take several tens of 
millions of years to transform to eclogite, and based on this 
suggestion Vorder Bruegge and Head [5] proposed that Maxwell 
Montes are 65 m.y. old. However, the recent crater distribution 
obtained from Magellan data suggests that the average age of Ishtar 
Terra is similar to that elsewhere on Venus, 500 m.y. [12,13]. To 
assess the effects of the lime lag in the phase change, the crustal 
thickening is halted at 120 m.y. and the crust is allowed to reach 
thermal equilibrium for the next 80 m.y. The temperature increase 
does not significantly reduce the volume proportion of eclogite. 


Fig. 1. Thermal evolution of the cold Venus model with a thickening crusL 
The numbers on Ihe curves are temperatures in °C. Figure lb shows the 
existing phases. = undepleied peridotile, 1 = depleted peridotite, 2 = basaltic 
crust, 3 = granulite, and 4 - eclogite. 

The surface topography produced by crustal thickening is deter- 
mined assuming Airy isostasy with a compensation depth at 150 km 
as suggested for the westempart of Ishtar Terra [14]. Figure2 shows 
the resulting topography with (curve 1 ) and without (curve 2) taking 
into account the BGET. The density of basalt is 2900 kg/m 3 at room 
temperature. That of granulite increases linearly and reaches -3500 
kg/m 3 when eclogite appears. Also taken into account is the density 
decrease with temperature. The constant density model in Fig. 2 is 
similar to that of B indschadler et al. 's [8] at steady -state condition, 
taking into account the density differences of the two models. 
However, in a more realistic model with the BGET, the surface 
topography attains a maximum of 1.8 km with a total crustal 
thickness of 38 km, beyond which the topography decreases due to 
sinking of the denser assemblages into the mantle. Halting the 
crustal thickening causes a rebound of the crust, but not enough